Tianyi Zhu

🔗 Simulate PyTorch Models with Simulink! 🚀 Engineers, unlock new possibilities by simulating PyTorch and other Python-based models directly in Simulink using co-execution blocks. This integration streamlines your workflow, enabling seamless model testing and validation. 🔍 What's Inside: >> Co-execution of Python models in Simulink << >> Enhanced simulation and validation processes << 👉 Dive into the details: https://spr.ly/6041o1cOX #DeepLearning #Simulink #Python #ModelSimulation #Engineering #MATLAB #MAthWorks #PyTorch

9

Senior Hannah Carroll in BME applied several machine learning models to images of retinas in order to score severity of diabetic retinopathy. To do this, she rapidly leveraged Pytorch and several computer vision deep learning neural net architectures this term in Deep Learning for Biomedical Engineering in UVM College of Engineering and Mathematical Sciences. GitHub: https://lnkd.in/gnrZkDY8

12

The parameters of compiler optimizations interact in complex and non-trivial ways, and finding optimal combinations remains an open research challenge. There are several approaches to discovering effective combinations of these optimization parameters. Recently, Michael Canesche (Cadence Design Systems) and Gaurav Verma (Stony Brook University) have had promising results using auto-tuning with coordinate descent. To visualize this process, imagine that each optimization represents a 'dimension' in a search space. The combination of all optimizations and their respective parameters forms a vector in this space. Now, consider that each point in this space is mapped to a real number representing the performance of a kernel (assuming a fixed input). Coordinate descent is a search technique that assumes the performance curve is convex. Based on this assumption, it aims to find the optimal point by varying one optimization parameter at a time. Explore as a Storm, Exploit as a Raindrop: On the Benefit of Fine-Tuning Kernel Schedulers with Coordinate Descent: https://lnkd.in/dgxxfdYR The Droplet Search Algorithm for Kernel Scheduling: https://lnkd.in/dBfQid9z

77

New repository for applying Constrained Deep Learning! Check out this blog post to learn how to design constrained models, and how constrained deep learning makes neural networks easier to verify 👉 https://spr.ly/6045gmKQv

206

1 Comment

Check out our latest paper, "Portability and Scalability Evaluation of Large-Scale Statistical Modeling and Prediction Software through HPC-Ready Containers," in Future Generation Computer Systems (The second most influential publication venue in the “Engineering & Computer Science —> Computing Systems” category for the 2024 Google Scholar Metrics! https://lnkd.in/ezUzCV9H ). This study is a joint work between the KAUST ECRC and Barcelona Supercomputing Center (BSC) and presents a dynamic solution for enhancing the portability and scalability of HPC applications. Learn how our (HPC-ready) containerization method, leveraging tools like #Spack and #Docker buildx, can maintain high performance with today's HPC software across diverse architectures, including cutting-edge GPU-accelerated systems. Our comprehensive scalability tests on #MareNostrum-4 demonstrate near-native performance and broad applicability to various HPC applications. Joint work with Jorge Ejarque (BSC), Omar Marzouk (Brightskies), Hatem Ltaief (KAUST), Rosa M. Badia (BSC), Marc Genton (KAUST), Ying Sun (KAUST), and David Keyes (KAUST). 📘 Read More: https://lnkd.in/epEsgPJS #HPC #HighPerformanceComputing #Containerization #Portability

28

3 Comments

The memory game: How to optimize memory for large language model inference and fine-tuning. Memory requirements for large language models (LLMs) can be daunting, but techniques like quantization, tensor parallelism, and model parallelism can help reduce memory usage while maintaining accuracy. Quantization, or using lower precision data types, can drastically reduce memory usage for LLM inference, allowing models to run on consumer GPUs. But it's not just about inference - fine-tuning LLMs for training requires even more memory. LLMem is a tool that accurately estimates memory usage for distributed fine-tuning methods across multiple GPUs. How can we continue to optimize memory for efficient LLM training and deployment? --- Hi, 👋🏼 my name is Doug, I love AI, and I post content to keep you up to date with the latest AI news. Follow and ♻️ repost to share the information! #largelanguagemodels #memoryoptimization #AImemoryusage

1

Mixture-of-Agents as an Event-Driven System 🤖☎️ This paper by Wang et al. shows you how to ensemble smaller LLMs to create a system that can outperform state-of-the-art larger models. We’ve implemented this paper in a fully async, event-driven workflow system thanks to Ravi Theja Desetty - treat each “small LLM” as an event-driven step that can process incoming events and respond to it, independently and in parallel. ✅ Take full advantage of processing an entire batch of requests  ✅ Cleaner, readable code LlamaPack: https://lnkd.in/gRuqQGk2 Learn more about workflows: https://lnkd.in/gqWwz7Xs

311

9 Comments

Does CPU speed significantly affect GPU inference in LLMs like llama.cpp? The answer may surprise you. 🔹 Our benchmarks with RTX 4080/4090 reveal CPU speed’s limited role in smaller model inference. 🔹 Intel vs. AMD showed comparable results, meaning fewer worries about CPU choice! 🔹 Key takeaway? Focus GPU inference optimizations elsewhere. Get the full benchmarking breakdown in our latest article. https://hubs.ly/Q02SrSVb0

5

Principal Component Analysis on manifolds? Learn how to approximate PCA on a manifold using Eigen decomposition (orthonormal basis) on the tangent space. #machinelearning #manifold #geometriclearning #PCA #ai

21

1 Comment

Large models can often overwhelm GPU memory, making inference slow & inefficient😓 In this blog post, Aman Juneja shares how to run tasks simultaneously across multiple processing units to reduce inference time without sacrificing accuracy 🎯 https://lnkd.in/gq5i8QpQ

3

👉 Learn how to simulate pre-trained models from #PyTorch and #TensorFlow directly in #Simulink with the new co-execution Simulink blocks. https://spr.ly/6044W3drs

2

🤖 Learn how to build a state-of-the-art chatbot application optimized for mobile processors, using tools like the OpenVINO™ Toolkit and PyTorch with Paula Ramos, PhD during her session at #AllThingsOpen happening today at 1:45pm.

42

Combining the power of TensorFlow, NeuralForecast, and zero-shot LLMs, Jinhang Jiang shows how we can generate accurate multi-step quantile forecasts.

11